INDEX

Explanations

technical terms and concepts related to programming and technology

oai_token-act-pair · gpt-4o-mini Triggered by @bot

moralize, gamify, lowball

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 16-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_65k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.16.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 iſt

-0.86

ſelf

-0.74

 Theſe

-0.74

 auffi

-0.72

 becauſe

-0.72

 Majefty

-0.71

^(@)

-0.69

 Beſ

-0.67

 Monfieur

-0.66

 deforma

-0.66

POSITIVE LOGITS

wise

0.54

 them

0.53

 образом

0.51

ally

0.51

 wise

0.47

lessly

0.47

matically

0.46

 istilah

0.46

ly

0.46

 Bourgoin

0.46

Activations Density 0.248%

technical terms and concepts related to programming and technology

moralize, gamify, lowball

No Comments

No Known Activations